OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.1

[der_wasi]

I'm using cgminer with a small number of 2pac bitcoin miners. In the USB menu there are a couple settings for Unplug and Reset. They don't seem to do anything to the miners. Am I missing something?

Thanks.

I´m not sure but maybe the miners don´t support that commands.

Hi

Is there any way to obtain a unique ID of the miner using the CGMiner API? Or at least the Mac Address of the network interface?

In BMMiner I can see a field called miner_id into [STATS0] but not in CGMiner. And I think it would be a good feature. Some times miners change the IP address if you are using DHCP. It will help to keep track of the miner.

Thanks!

Why do you use DHCP in the first place? Set the device you´r mining with to a static IP.

[jdbravo]

With many miners it may be a pain... with DHCP it's much easier

[der_wasi]

What do you mean by "miners"? What hardware do you use?

[sabre_labs]

Guys,

Technical question: How does CGminer split its job in an ASIC miner?

Background (I tried to go in depth in the C code but I still can't find the right function for it):
- I am trying to hack together a ASIC miners' hashing boards to work together in a particular way
- Assuming that it's for Bitcoin mining and given an ASIC miner like an Antminer S9 etc and assuming I have the drivers for it
- I know that Stratum will issue the extranonce2_size and other header details to CGminer
- How does CGminer split the job between the different boards/chips? I assume it splits the job based on the returned extranonce2 values generated as it doesn't look like a simple increment of the extranonce2 from 0x00 up.

Any ideas or fingers pointing in the right direction will be super appreciated.

Thanks all

[-ck]

Guys,

Technical question: How does CGminer split its job in an ASIC miner?

Background (I tried to go in depth in the C code but I still can't find the right function for it):
- I am trying to hack together a ASIC miners' hashing boards to work together in a particular way
- Assuming that it's for Bitcoin mining and given an ASIC miner like an Antminer S9 etc and assuming I have the drivers for it
- I know that Stratum will issue the extranonce2_size and other header details to CGminer
- How does CGminer split the job between the different boards/chips? I assume it splits the job based on the returned extranonce2 values generated as it doesn't look like a simple increment of the extranonce2 from 0x00 up.

Any ideas or fingers pointing in the right direction will be super appreciated.

This line:
https://github.com/ckolivas/cgminer/blob/master/cgminer.c#L2415

[sabre_labs]

This line:
https://github.com/ckolivas/cgminer/blob/master/cgminer.c#L2415

Thanks CK!

[EarthWide]

I finally updated myself from an older version of CG miner to the current version for my U1.  When I used the start command line of

cgminer.exe -o stratum+tcp://pool.ckpool.org:3333 -u 12fvYKMQ7BgBuWqVfTvq8VH1TKnh7zNB3z.USB1 -p  --anu-freq 275 --suggest-diff 20

CGminer automatically picked it up as a AU3 not the ANU that it should have.  After a couple restarts it went back to recognizing it at an ANU.  No real issue, but when it was thinking it was a AU3 I was getting some weird error messages. 

Is there a way to force CGminer to only stick to ANU recognition?

[-ck]

Is there a way to force CGminer to only stick to ANU recognition?

Not really. There was hardly anything different between many many devices that used the exact same communication protocol because hardware manufacturers didn't want to think about it, used the same usb chip, and were too lazy to add a simple identifier on their usb devices. Adding more commands now for such old slow hardware is a bit of a waste of time I'm afraid.

[EarthWide]

Is there a way to force CGminer to only stick to ANU recognition?

Not really. There was hardly anything different between many many devices that used the exact same communication protocol because hardware manufacturers didn't want to think about it, used the same usb chip, and were too lazy to add a simple identifier on their usb devices. Adding more commands now for such old slow hardware is a bit of a waste of time I'm afraid.

Totally understandable!  Thanks for the very quick reply CK!

Edit:
This is the error message that gets shown:
[2018-02-07 17:44:01] AU3 0: No valid hashes for over 2 secs, attempting to reset
 [2018-02-07 17:44:01] AU3 0 attempted reset got err:(0) LIBUSB_SUCCESS
 [2018-02-07 17:44:05] AU3 0: No valid hashes for over 2 secs, attempting to reset
 [2018-02-07 17:44:05] AU3 0 attempted reset got err:(0) LIBUSB_SUCCESS

I don't expect any update or changes, I only value my U1 as a realtivly cheap tool to get my training wheels off.

[Jay_Pal]

I finally updated myself from an older version of CG miner to the current version for my U1.  When I used the start command line of

cgminer.exe -o stratum+tcp://pool.ckpool.org:3333 -u 12fvYKMQ7BgBuWqVfTvq8VH1TKnh7zNB3z.USB1 -p  --anu-freq 275 --suggest-diff 20

CGminer automatically picked it up as a AU3 not the ANU that it should have.  After a couple restarts it went back to recognizing it at an ANU.  No real issue, but when it was thinking it was a AU3 I was getting some weird error messages. 

Is there a way to force CGminer to only stick to ANU recognition?

Sometimes it happens.
Usually mine start to misbehave as AU# instead of ANU#, but allways power up as ANU#
I've even downgraded to 4.9.2 since they were quite unstable on 4.10.0.
I have mine at 261 but even at 255 they become zombies - do yours run stable at 275?!

[Gavri1a]

Hi there
could you tell me how to use this key?

--disable-rejecting

sometimes I get the 'rejected' error from pool and then after a while I get the message 'your IP is banned'

I'd like cgminer to connect to another pool for 10 minutes (m.b. 20)
and after 10 minutes left, reconnect to the first pool

is that possible to set up?

[-ck]

Hi there
could you tell me how to use this key?

--disable-rejecting

sometimes I get the 'rejected' error from pool and then after a while I get the message 'your IP is banned'

I'd like cgminer to connect to another pool for 10 minutes (m.b. 20)
and after 10 minutes left, reconnect to the first pool

is that possible to set up?

Being rejected by the pool is different to getting all rejected shares which is what that is for. There is no command for what you're asking. If you're being rejected by the pool you are having a configuration issue with your pool setup, or your pool is shit.

[Dkristof]

Does not work on Windows 7 x64
error: missing api-ms-win-core-libraryloader-l1-2-0.dll
WTF?

[aldoyh]

All of the sudden the is inactive, although it's listed as shown in the attached, how do I fix this?

https://imgur.com/a/8aeO0

[-ck]

Is it relevant to mine with 1 home PC? and tell me please, what actually currency we can mine sorry, if my questions are newvbie)

Did you try reading the mining section introduction?
https://bitcointalk.org/index.php?topic=2415854.0

[alexglvr]

Hello,

I installed my own bitcoin mining pool using unomp (ubuntu VM) and I now try to run cgminer to test GPU mining on it.

I am new in mining, and do not manage to make cgminer run...

Here is my actual .bat

Code:

DEL *.BIN
CLS
@ECHO ON
ECHO.
setx GPU_MAX_ALLOC_PERCENT=100
setx GPU_USE_SYNC_OBJECTS=1

timeout /t 4

cgminer.exe --scrypt

and my cgminer.conf file :

Code:

{
"pools" : [
	{
		"url" : "stratum+tcp://192.168.1.13:3032",
		"user" : "toto",
		"pass" : "foobar"
	}
]
,
"scrypt" : true,
"intensity" : "20",
"vectors" : "1",
"worksize" : "256",
"kernel" : "scrypt",
"lookup-gap" : "0",
"thread-concurrency" : "24550",
"shaders" : "0",
"gpu-engine" : "0",
"gpu-fan" : "100",
"gpu-memclock" : "0",
"gpu-memdiff" : "0",
"gpu-powertune" : "0",
"gpu-vddc" : "0.000",
"temp-cutoff" : "95",
"temp-overheat" : "90",
"temp-target" : "85",
"api-mcast-port" : "4028",
"api-port" : "4028",
"expiry" : "120",
"failover-only" : true,
"gpu-dyninterval" : "7",
"gpu-platform" : "0",
"gpu-threads" : "1",
"hotplug" : "5",
"log" : "10",
"no-pool-disable" : true,
"per-device-stats" : true,
"queue" : "1",
"scan-time" : "60",
"temp-hysteresis" : "3",
"shares" : "0",
"kernel-path" : "/usr/local/bin"
}

But I got errors :
 
[2018-02-20 13:16:14] Init GPU thread 0 GPU 0 virtual GPU 0
 [2018-02-20 13:16:14] CL Platform vendor: NVIDIA Corporation
 [2018-02-20 13:16:14] CL Platform name: NVIDIA CUDA
 [2018-02-20 13:16:14] CL Platform version: OpenCL 1.2 CUDA 9.1.92
 [2018-02-20 13:16:14] List of devices:
 [2018-02-20 13:16:14]  0       GeForce GTX 1050 Ti
 [2018-02-20 13:16:14] Selected 0: GeForce GTX 1050 Ti
 [2018-02-20 13:16:14] Selecting scrypt kernel
 [2018-02-20 13:16:14] Initialising kernel scrypt130511.cl without bitalign, 1 vectors and worksize 256
 [2018-02-20 13:16:14] Maximum buffer memory device 0 supports says 1073741824
 [2018-02-20 13:16:14] Your scrypt settings come to 1608908800
 [2018-02-20 13:16:14] initCl() finished. Found GeForce GTX 1050 Ti
 [2018-02-20 13:16:14] 1 gpu miner threads started
 [2018-02-20 13:16:14] Work update message received
 [2018-02-20 13:16:14] Network diff set to 3.01T
 [2018-02-20 13:16:14] New block: 0000000000000000005cabc1047747a12e81bcd18bdb7e8e41e6a4a1ed450244... diff 3.01T
 [2018-02-20 13:16:16] Error: clEnqueueReadBuffer failed error -5. (clEnqueueReadBuffer)
 [2018-02-20 13:16:16] GPU 0 failure, disabling!
 [2018-02-20 13:16:24] ALL (10s):0.000 (avg):0.000h/s | A:0  R:0  HW:0  WU:0.0/m
 [2018-02-20 13:16:32] Work update message received
 [2018-02-20 13:16:34] ALL (10s):0.000 (avg):0.000h/s | A:0  R:0  HW:0  WU:0.0/m

I use CGMiner 3.7.2 and I have a NVIDIA 1050 TI

What did I miss?

Thank you for your help,

Alex

[-ck]

I installed my own bitcoin mining pool using unomp (ubuntu VM) and I now try to run cgminer to test GPU mining on it.

I use CGMiner 3.7.2 and I have a NVIDIA 1050 TI

Unsupported. Only the current version and bitcoin mining with ASICs is supported here.

[kenkos68]

I hope this isn't a dumb question...

Going through some of my debug logs, I noticed that the generated stratum header has the version like this:

[2018-02-15 23:08:23.772] Generated stratum header 20000000ed37665c908b8f6b9120d09918afa1ef1c3bd3fa0024d3f700000000000000005fb360c99001606 f4f9aeb9293c5776d6e3e666adf0e6dacdbb9526ae3561fa95a86592f1761e9f800000000000000 8000000000000000000000000000000000000000000000000000000000

Shouldn't it be like this?:

[2018-02-15 23:08:23.772] Generated stratum header 00000020ed37665c908b8f6b9120d09918afa1ef1c3bd3fa0024d3f700000000000000005fb360c99001606 f4f9aeb9293c5776d6e3e666adf0e6dacdbb9526ae3561fa95a86592f1761e9f800000000000000 8000000000000000000000000000000000000000000000000000000000

I didn't see anywhere in the code where the bbversion was reversed.  Unless I missed it.  I did see some previous posts from way back where the block version is shown in the logs like my second example.  

The version gets put in the header_bin as in the first example, not reversed.    Raw block headers have the version reversed and when test hashed, produce the correct hashes.

Thanks,

-Ken

[-ck]

I hope this isn't a dumb question...

Going through some of my debug logs, I noticed that the generated stratum header has the version like this:

[2018-02-15 23:08:23.772] Generated stratum header 20000000ed37665c908b8f6b9120d09918afa1ef1c3bd3fa0024d3f700000000000000005fb360c99001606 f4f9aeb9293c5776d6e3e666adf0e6dacdbb9526ae3561fa95a86592f1761e9f800000000000000 8000000000000000000000000000000000000000000000000000000000

Shouldn't it be like this?:

[2018-02-15 23:08:23.772] Generated stratum header 00000020ed37665c908b8f6b9120d09918afa1ef1c3bd3fa0024d3f700000000000000005fb360c99001606 f4f9aeb9293c5776d6e3e666adf0e6dacdbb9526ae3561fa95a86592f1761e9f800000000000000 8000000000000000000000000000000000000000000000000000000000

I didn't see anywhere in the code where the bbversion was reversed.  Unless I missed it.  I did see some previous posts from way back where the block version is shown in the logs like my second example.  

The version gets put in the header_bin as in the first example, not reversed.    Raw block headers have the version reversed and when test hashed, produce the correct hashes.

Thanks,

-Ken

Take a look at some blocks to see what version looks like these days, eg:
https://btc.com/00000000000000000077140243064000ba35b22d3e2440e936a722b90cd09365

Compare with older blocks, eg:
https://btc.com/00000000000000001ed61d4b7dd337c8eb1de0f21a0bd310a8733fb35f21991c

[kenkos68]

Thanks ck!

My question was geared more towards how the version is placed into header_bin, pool->header, and eventually work->data prior to being hashed.

The older block you listed gets hashed as:  https://blockchain.info/block/00000000000000001ed61d4b7dd337c8eb1de0f21a0bd310a8733fb35f21991cformat=hex
02000000ea7ee4542a0d7496954302e67b99a4d1a3d1c2724f8ddb1a00000000000000007d28ed3cb6e06dc df8267afaa908999d48f2d4deba94b4e344c3e126a3c8b00b174e1f54e9db2418e5bc702

The newer as (and current blocks):  https://blockchain.info/block/00000000000000000077140243064000ba35b22d3e2440e936a722b90cd09365?format=hex
00000020633a9073c5b1a5f32afefa4c1116c5bb9c593dfa419812000000000000000000515b81eeca6c575 93c90264ee1a1cb581fc3b7052e8adc23ec16db9917cf828bd33d575ac19100184dbccd28

Current blocks manually double sha256 hashed give correct results with the second format.

I guess I am not seeing in the code where the version is flipped after it is placed in as "20000000" prior to being hashed.

Thanks again for your help!

-Ken

Take a look at some blocks to see what version looks like these days, eg:
https://btc.com/00000000000000000077140243064000ba35b22d3e2440e936a722b90cd09365

Compare with older blocks, eg:
https://btc.com/00000000000000001ed61d4b7dd337c8eb1de0f21a0bd310a8733fb35f21991c